Agent Types and Capabilities

Clear overview of what agent types exist in Ona and what each can do.

This is the fastest reference for:

  • which agent handles what
  • what capabilities are available
  • where boundaries and limits apply

Quick Model

  • User talks to Solin
  • Solin delegates to specialist agents
  • Specialist agents execute tasks and tools
  • Oversight/security agents monitor and enforce safety

Core Agent Types

Agent Type Primary Purpose Key Capabilities Limits
Solin (Squad Lead) Main orchestrator and user-facing assistant Understand mission intent, split work into tasks, delegate to specialists, combine final answer Does not execute tool loop directly in full-job mode
Task Specialist Agents Execute domain work (writing, research, code, design, etc.) Run tools, use memory, search, call assistants/subordinates, produce artifacts Scoped to assigned tasks and policies
Observer (optional) Mission quality/safety oversight Watch mission flow, check on-task behavior, flag risky outputs, report to Solin Does not execute normal mission tasks
Guardian (optional) Security enforcement Validate tool params/outputs, detect attacks/misuse, block high-risk actions, report violations Does not execute normal mission tasks
Operations Agents (optional) Domain-specific operations (secrets/certs/project workflows) Lockbox operations, certificate/key lifecycle support, project/task operations Limited by approvals, policies, and enabled features
Assistant/Subordinate Agents Extra capacity and specialization Parallelize subtasks, support specialist workload, role-based execution Ephemeral by default; bounded to delegated scope

Specialist Agent Personas

These are common role templates Solin can use for execution:

  • Writer
  • Content Creator
  • Social Content Creator
  • Researcher
  • Developer
  • Designer
  • SEO
  • Marketer
  • Sales Agent
  • Landing Page
  • Project Agent

You can also create your own custom agents based on your own requests (for example a niche role for your workflow), and Solin can include them in mission delegation when configured.

Typical capability patterns

Persona Group Typical Capabilities
Content (Writer/Content/Social/Marketing/Sales) Drafting, campaign assets, structured messaging, outreach content
Research (Researcher/SEO) Discovery, synthesis, keyword/structure recommendations, evidence-backed summaries
Build (Developer/Designer/Landing Page) Code and implementation tasks, UI/design iteration, landing/funnel output, build/test-oriented execution
Operations (Project Agent) Project/task tracking, status management, execution planning support

Oversight and Security Agents

Observer Agent

Focus:

  • mission alignment checks
  • quality and safety watch
  • anomaly signaling to Solin

Observer can:

  • read mission activity context
  • report concerns in mission flow
  • help reduce off-track execution

Observer cannot:

  • replace specialist execution
  • bypass approval/security controls

Guardian Agent

Focus:

  • policy/security enforcement
  • prompt/tool abuse detection
  • risk blocking before execution

Guardian can:

  • inspect risky tool patterns
  • enforce deny/warn behavior based on policy
  • escalate or block high-risk operations

Guardian cannot:

  • act as a replacement for business/task specialists
  • bypass governance boundaries

Operations-Focused Agents

Lockbox Agent

Primary purpose:

  • manage lockbox lifecycle operations (add/list/remove key IDs through controlled flow)

Capabilities:

  • supports secure credential management workflows
  • works with approval flow for sensitive actions

Limits:

  • should never expose plaintext secrets in responses

Certificate Manager

Primary purpose:

  • certificate and key lifecycle operations

Capabilities:

  • status/renewal/rotation-oriented workflows
  • supports operational certificate health management

Limits:

  • follows approval and security policy requirements

Project Agent

Primary purpose:

  • project and task management support

Capabilities:

  • create/list/update project structures
  • add/track operational tasks

Limits:

  • not a replacement for domain specialists (research/code/design)

Interaction Modes and Capability Depth

Interaction Path What Usually Happens Capability Depth
Quick reply Fast response path for simple interactions Lightweight, low-latency response
Full mission job Solin orchestrates specialist agents + tools Full multi-step execution and collaboration

Use full mission mode when you need deeper output, multi-step work, or specialized execution.


Boundaries That Always Apply

  • user-facing conversation goes through Solin
  • specialist agents execute within assigned mission scope
  • approvals are required for sensitive actions
  • oversight/security controls can intervene in high-risk flows
  • memory/tool behavior follows configured policy and scope

Related Docs